File size: 8,186 Bytes
4460c42
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyNAxD9Hy7SaN4kD/p7d0PC5",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/palbha/airline-faq-rag/blob/main/faq_data_generator.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Install necessary libraries if required - This code ran on Google Colab & the libraires where supported by default- please rephrase this\n",
        "\n"
      ],
      "metadata": {
        "id": "-Ruq0mXsA9do"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import csv\n",
        "import json\n",
        "import os\n",
        "import openai\n",
        "import csv\n",
        "from openai import OpenAI\n",
        "from google.colab import userdata\n",
        "\n",
        "#Based on airlines FAQ I identified potential topics which can be shared with our agent to create FAQ's\n",
        "FAQ_TOPICS = [\n",
        "    \"Airport Services\",\n",
        "    \"Animal Transportation\",\n",
        "    \"Beyond Business\",\n",
        "    \"Booking and managing a reservation\",\n",
        "    \"Carbon Offsetting\",\n",
        "    \"AirlineX Compliance\",\n",
        "    \"Hotels, cars and travel insurance\",\n",
        "    \"AirlineX Offers\",\n",
        "    \"On-board experience\",\n",
        "    \"Operational Updates\",\n",
        "    \"Payments\",\n",
        "    \"Privilege Club : Qatar Airways' loyalty programme\",\n",
        "    \"ArlineX Airways Affiliate Program\",\n",
        "    \"ArlineX Airways Packages\",\n",
        "    \"ATravel - ArlineX Loyalty Program\",\n",
        "    \"ATravel - ArlineX Loyalty Program - Account Cancellation\",\n",
        "    \"ATravel - ArlineX Loyalty Program - Account Management\",\n",
        "    \"ATravel - ArlineX Loyalty Program - Booking Terms and Conditions\",\n",
        "    \"Travel Baggage\",\n",
        "    \"Baggage\",\n",
        "    \"BAGTAG\",\n",
        "    \"Hand baggage\",\n",
        "    \"Liquids\",\n",
        "    \"Mishandled baggage\",\n",
        "    \"Travel voucher\",\n",
        "    \"Voucher redemption\",\n",
        "    \"TripAdd\",\n",
        "    \"eSIM - TripAdd\",\n",
        "    \"Lounge - TripAdd\",\n",
        "    \"Meet and Greet - TripAdd\",\n",
        "    \"Young Travellers\",\n",
        "    \"Travelling with children\",\n",
        "    \"Unaccompanied minors\",\n",
        "]"
      ],
      "metadata": {
        "id": "aAvEjDrrRNJE"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "openai = OpenAI(\n",
        "   base_url=\"https://generativelanguage.googleapis.com/v1beta/\",\n",
        "  api_key=userdata.get('gemini_api'),\n",
        ")"
      ],
      "metadata": {
        "id": "SWnjOAg7BeO_"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def get_faq_ques_for_topic(topic):\n",
        "    messages = [\n",
        "        {\n",
        "            \"role\": \"system\",\n",
        "            \"content\": (\n",
        "                \"\"\"You are an assistant that generates FAQ-style questions  for an airline named Airline X, which operates international and domestic flights in Canada.\n",
        "\n",
        "For each topic, Generate  realistic and informative user-style questions for the FAQ topic. Do no include answers\n",
        "\"\"\"\n",
        "            )\n",
        "        },\n",
        "        {\n",
        "            \"role\": \"user\",\n",
        "            \"content\": f\"Generate FAQ 5-10 questions  about the topic: '{topic}'.\"\n",
        "        },\n",
        "        {\n",
        "            \"role\": \"system\",\n",
        "            \"content\": \"Return ONLY a valid JSON array of question objects. Do not include answers\"\n",
        "        }\n",
        "    ]\n",
        "    response = openai.chat.completions.create(\n",
        "        model=\"gemini-1.5-flash\",\n",
        "        messages=messages,\n",
        "        temperature=0.7,\n",
        "        max_tokens=700,\n",
        "    )\n",
        "    # The model response should be a JSON array of objects like: [{\"question\": \"...\", \"answer\": \"...\"}, ...]\n",
        "    content = response.choices[0].message.content.strip()\n",
        "\n",
        "\n",
        "    return content\n",
        "\n"
      ],
      "metadata": {
        "id": "64J-8B5hRSMf"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def get_faq_ans_for_topic(topic,question):\n",
        "    messages = [\n",
        "        {\n",
        "            \"role\": \"system\",\n",
        "            \"content\": (\n",
        "                f\"You are an assistant that generates FAQ-style answers for an airline named Airline X, \"\n",
        "        f\"which operates international and domestic flights in Canada.\\n\\n\"\n",
        "        f\"The FAQ topic is: '{topic}'.\\n\"\n",
        "\n",
        "            )\n",
        "        },\n",
        "        {\n",
        "            \"role\": \"user\",\n",
        "            \"content\": f\"Generate FAQ answers for the question: '{question}'.\"\n",
        "        }       ,\n",
        "        {\n",
        "            \"role\": \"system\",\n",
        "            \"content\": \"Provide a clear, self-contained, and factual-sounding answer based on Airline X's own policies. \"\n",
        "        \"Do NOT reference any website, support, or external links. Make the answer complete, realistic, \"\n",
        "        \"and independent of outside context. Return ONLY answers\"\n",
        "        }\n",
        "    ]\n",
        "    response = openai.chat.completions.create(\n",
        "        model=\"gemini-1.5-flash\",\n",
        "        messages=messages,\n",
        "        temperature=0.7,\n",
        "        max_tokens=700,\n",
        "    )\n",
        "    # The model response should be a JSON array of objects like: [{\"question\": \"...\", \"answer\": \"...\"}, ...]\n",
        "    content = response.choices[0].message.content.strip()\n",
        "\n",
        "\n",
        "    return content\n",
        "\n"
      ],
      "metadata": {
        "id": "wZhnhpHLSi7c"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "faq_data = []\n",
        "import re\n",
        "for topic in FAQ_TOPICS:\n",
        "    questions=get_faq_ques_for_topic(topic)\n",
        "    raw_text = questions.strip()\n",
        "    cleaned = re.sub(r\"^```json|```$\", \"\", raw_text, flags=re.IGNORECASE).strip(\"`\\n \")\n",
        "\n",
        "    # Now attempt to parse\n",
        "    question_data = json.loads(cleaned)\n",
        "    for key in question_data:\n",
        "      answer=get_faq_ans_for_topic(topic,key['question'])\n",
        "      faq_data.append({\n",
        "            \"topic\": topic,\n",
        "            \"question\": key['question'],\n",
        "            \"answer\": answer\n",
        "        })\n"
      ],
      "metadata": {
        "id": "bzTtCQijTlmq"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "pd.DataFrame(faq_data).to_csv(\"faq_data.csv\",index=False)"
      ],
      "metadata": {
        "id": "nKwJfsMEWstr"
      },
      "execution_count": 7,
      "outputs": []
    }
  ]
}